Artificial Intelligence in Oncology - Supporting scientific research
Radboudumc
In short, Kalijn Bols' research entails the following:
'Uveal melanoma occurs only in ~200 patients a year in the Netherlands and up to half of the patients with a primary uveal melanoma will develop metastases. Uveal melanoma differs clinically and genetically from cutaneous melanoma and response rates to immunotherapy are low. Previously, it was believed that the development of UM at an immune-privileged site, the eye, but infiltration of immune cells in uveal melanoma is present. In this project, we will analyze multiplex immunohistochemistry imaging and flow cytometry data of multiple patient cohorts to investigate tumor-immune interactions. To avoid the risk of overfitting our real data and overcome the challenges with multimodal data fusion, we will first develop and optimize our analysis approach using simulated data resembling real data. By optimal analysis of the combined data from multiple heterogeneous sources, we aim to obtain mechanistical understanding of immunological processes which will help develop strong theoretical rationales for clinical trials in the future to improve the outcome of uveal melanoma patients.'
We have expanded our data on immune cells in the tumor microenvironment in uveal melanoma. Tissue slides are stained, digitally imaged and analyzed using our custom-made neural network. We are also using advanced methodology for the fusion of clinical data from heterogeneous cohorts.
We have expanded our data on immune cells in the tumor microenvironment in uveal melanoma. Tissue slides are stained, digitally imaged and analyzed using our custom-made neural network.